Yes, this is a good point. Hiring "the best" is more like a multi-variable optimization problem. Pareto showed mathematically its very unlikely that there is

Yes, this is a good point. Hiring "the best" is more like a multi-variable optimization problem. Pareto showed mathematically its very unlikely that there is a single "best", instead there is a population of non-dominated solutions that form the "Pareto frontier" which is a population of "less bad than other" choices. One variable in that might be gender. But again, forming a list of female engineers seems like its premature optimization on a single axis. The best approach in these cases is as I said on another comment, take your top candidates, some are likely to be female and you can weight that dimension fittingly (e.g. we have no women, its slightly more important than having just any another C++ dev).